[Show abstract][Hide abstract] ABSTRACT: The diminished accuracy of port-based and payload-based classification motivates use of transport layer statistics for network traffic classification. A semi-supervised clustering approach based on improved K-Means clustering algorithm is proposed in this paper to partition a training network flows set that contains a huge number of unlabeled flows and scarce labeled flows. The variance of flow attributes is used to initialize clusters centers instead of the random selection of the cluster centers in initialization. Scarce labeled flows are selected to construct a mapping from the clusters to the predefined traffic classes set. The experimental results show that both the overall accuracy and square error (SSE) value of our algorithm present better than those based on normal K-Means algorithm defined in Ref. .
No preview · Article · Dec 2010 · The Journal of China Universities of Posts and Telecommunications
[Show abstract][Hide abstract] ABSTRACT: Being a novel category of watermarking schemes, reversible watermarking algorithms were developed in recent years. As it can recover the watermarked data back to the original host signal, reversible watermarking algorithms are suitable for medical, military and other special fields. However, these algorithms have their defects, such as weak robustness, low embedding capacity and high calculating complexity. This paper proposes a near reversible watermarking algorithm based on LSB replacement. It can not only recover the original data to a high extent, but also have strong robustness and low calculating complexity.